JAK: Not Just Another Kinase

Janus kinase or “just another kinase” (JAK) is a family of intracellular, non-receptor tyrosine kinases that transduce cytokine-mediated signals via the JAK–STAT pathway (1, 2). The name Janus originates from the two-faced Roman god of duality, because JAKs possess two similar phosphate-transferring domains, one of which displays the kinase enzymatic activity, whereas the other motif negatively regulates the kinase activity of the first in a feedback loop (1).

JAK–STAT signaling is composed of three major proteins: Cell-surface receptors, JAKs, and STATs. JAKs function as tyrosine kinase enzymes that are bound to the cytoplasmic regions of type I and II cytokine receptors. Once a ligand binds to the receptor, JAKs add phosphates to the receptor. Two STAT proteins then bind to the phosphates and form a dimer. The dimer enters the nucleus, binds to DNA, and transcription of target genes ensues. Overall, the JAK–STAT pathway is vital for the proliferation and survival of cancer cells and may contribute to drug resistance (2).

The JAK–STAT pathway is activated by over 50 different pro-cytokine receptors that control hematopoiesis, the immune response, embryogenesis, and inflammation through the signaling pathway (3–6). This pathway has an important role in the pathogenesis of a variety of immune-mediated diseases and malignant processes.

JAK inhibitors are a class of medications that function to block signaling through the JAK–STAT pathway (3). JAK inhibitors (also called jakinibs) function by preventing the JAK protein from phosphorylating, therefore attenuating intracellular signaling. The most extensive clinical studies for JAK inhibitors have been in rheumatoid arthritis (RA) and myelofibrosis, but JAK inhibitors have also been applied in dermatology for treatment of conditions such as psoriasis/psoriatic arthritis, atopic dermatitis, alopecia areata, vitiligo, and dermatomyositis (5–22).

Over 50 cytokines signal via the JAK/STAT pathway to orchestrate hematopoiesis, regulate immunity, and induce inflammation. The main function of the JAK–STAT pathway is to transfer signals from the receptors on the cell membrane, which can be categorized as, including IL receptors, IFN receptors, or colony-stimulating factor receptors (CSFR), to the nucleus. The pathway is necessary for cytokines and growth factor function, leading to essential cellular processes such as myeloid and lymphoid differentiation/proliferation/hematopoiesis, lactation, development of the immune system, and proinflammatory response (Fig. 1; refs. 23–28).

Cytokines are secreted glycoproteins that operate as intercellular messengers, stimulating differentiation, proliferation, and programmed cell death of their target cells. They act by binding to specific receptors on the target cell surface and switching on a phosphotyrosine-based intracellular signaling cascade initiated by kinase enzymes and then propagated by SH2 domain-containing transcription factors. As cytokine signaling is proliferative and often inflammatory, its amplitude and duration are tightly controlled.

Most cytokines are small helical-bundle proteins, generally 150–200 amino acids in length (4). They are divided into two classes based on elements discerned in their receptors. Class I cytokines consist of four α-helices in an up–up–down–down configuration. Some of these, such as IL5, exist as dimers, but the topology is conserved. The up–up–down–down conformation requires two long loops to connect the up–up and down–down pairs. In class II cytokines, one or both of these loops is replaced by an extra α-helix resulting in five to six helices arranged in an anti-parallel manner.

Class I cytokines signaling through JAK–STAT include, but are not limited to IL2, IL3 family, IL4, IL6 family, IL7, IL9, IL12, IL13, IL15, IL21, IL23, G-CSF, GM-CSF, EPO, TPO (4). Class II cytokines include, but are not limited to IFN-alpha, -beta, -epsilon, -kappa, -omega, -gamma, IL10, IL19, IL20, IL22, IL24, IL26 (4).

The cytokine signaling cascade needs only three elements (receptor, kinase, and transcription factor) to elicit a response. Each cytokine binds to a specific receptor on its target cell surface. These receptors contain intracellular motifs that are constitutively associated with members of the JAK family of tyrosine kinase enzymes. The four JAK family members are: JAK1, JAK2, JAK3, and tyrosine kinase 2 (TYK2). JAKs are inactive before cytokine exposure. However, once the cytokine binds to its receptor, it induces their auto-activation by transphosphorylation. Once activated, JAKs phosphorylate the intracellular tails of the receptors on specific tyrosines, which in turn act as docking sites for member of the STAT family of transcription factors. The STAT family is composed of seven members STAT1, STAT2, STAT3, STAT4, STAT5a, STAT5b, and STAT6, which mainly act, as mentioned, as transcription factors. Receptor-localized STATs are then phosphorylated by JAK, which leads to their disassociation from the receptor and translocation to the nucleus, where they drive the expression of cytokine-responsive genes. To guarantee that signaling is properly switched off, a number of proteins act to dampen cytokine signaling at numerous levels of the pathway. Importantly, the suppressors of cytokine signaling (SOCS) family are negative feedback inhibitors of the signaling cascade. A general rule of cytokine signaling is that each cytokine binds to a specific receptor, which promotes activation of specific JAK(s) and STAT(s), and signaling is switched off by a specific SOCS protein.

JAK family members–JAK1, JAK2, JAK3, and TYK2—are implicated in cell growth, survival, and development, and are critically important for differentiation of a variety of cells, including immune cells and hematopoietic elements. Therefore, it is not surprising that germline mutations or specific polymorphisms in these genes are associated with disease. A striking phenotype associated with inactivating JAK3 germline mutations is severe combined immunodeficiency syndrome (Table 1; refs. 29–36). JAK3 missense mutations have also been found in patients with clear cell renal cell cancer. Pediatric acute lymphoblastic leukemia has been associated anecdotally with germline TYK2 mutations, and hereditary thrombocytosis with JAK2 germline mutations. Finally, the autoimmune disease systemic lupus erythematosus has been linked to TYK2 polymorphisms.

Both activating and loss-of-function somatic alterations in JAK genes can occur. Activating JAK alterations are associated with development of mostly hematologic malignancies. Loss-of function JAK2 alterations are associated with immune checkpoint blockade resistance (37–57).

Somatic alterations in various JAK genes are predominantly associated with a variety of hematologic malignancies (Table 2; refs. 37–57). As an example, the acquired somatic mutation JAK2 V617F, which constitutively activates JAK2 kinase, is the most frequent molecular event in myeloproliferative disorders. It is observed in about 95% of cases of polycythemia vera and in 55%–60% cases of essential thrombocytosis and primary myelofibrosis. JAK2 V617F is also found in the normal population. Among 49,488 individuals from the Copenhagen General Population Study, 63 (0.1%) tested positive for the JAK2V617F mutation. Of these, 48 were eventually diagnosed with a myeloproliferative neoplasm (49). JAK2 V617F mutations in healthy people are referred to as clonal hematopoiesis of indeterminate potential (CHIP), which defines the presence of a clonally expanded hematopoietic stem cell caused by a leukemogenic mutation in individuals without evidence of dysplasia, cytopenia or hematologic malignancy. CHIP correlates with a 0.5%–1.0% risk per year of leukemia. Importantly, CHIP also confers a 2-fold increase in cardiovascular risk independent of conventional risk factors and is also associated with venous thromboembolism and other inflammatory states. Other CHIP mutations (in addition to JAK2 V617F) occur in DNA damage repair genes PPM1D and TP53, epigenetic regulators ASXL1, DNMT3A, TET2, and mRNA spliceosome components SF3B1, and SRSF2, as well as other genes (50). Interestingly, patients with Erdheim Chester Disease, a non-Langerhans histiocytosis, appear to have elevated rates of myeloproliferative disorders, sometimes heralded by a JAK2 V617F mutation (51). JAK family mutations are also observed in a subset of patients with a variety of T-cell malignancies, including T-ALL, T-PLL, and Sezary syndrome, as well as childhood B-cell precursory ALL, and acute megakaryoblastic leukemia (with and without Down syndrome (37–48). An activating JAK1 mutation has been reported in Castleman's disease as well and may explain an exceptional response to the anti-IL6 antibody siltuximab in the reported patient in the absence of elevated IL6 levels, because the JAK1 mutation may have sensitized the receptor to the IL6 ligand (42). Finally, a TEL–JAK2 fusion protein that includes the catalytic domain of JAK2 and the TEL-specific oligomerization domain, resulting in constitutive activation of its tyrosine kinase activity, has been seen in ALL and in atypical chronic myelogenous leukemia (44).

Somatic genomic copy-number alterations, involving loss of the key IFNγ gene function inducer JAK2, are a pivotal driver of immune resistance and escape. Acquired resistance to immune checkpoint therapy, including PD-1 blockade, in patients with advanced melanoma and other cancers has been associated with JAK2 deletion and loss-of-function mutations, respectively (52–55). In depth study showed that the deletion of JAK2 and PD-LI, two neighboring genes found on chromosome 9p24, was associated with primary resistance to anti–PD-1 immunotherapy in recurrent human papillomavirus–negative head and neck squamous cancers (54). The latter findings regarding resistance to immune checkpoint blockade were confirmed (55), and CAP/CLIA validated for use in the clinic to select optimal therapy in this setting. Therefore, lack of JAK2-mediated IFNγ responsiveness allows cancer cells to escape from antitumor T cells and, in the context of anti–PD-1/L1 immunotherapy, abrogates the antitumor efficacy of this approach. Consistent with these PD-L1–JAK2 co-deletion resistance findings, JAK2, PD-L1 (and PDL2; 9p24.1) amplification, or 9p copy-number gain, have been associated with opposite effects, namely anti–PD-1 immunotherapy benefit. Increased PD-L1 expression, which can be directly caused by 9p24.1 amplification (or 9p copy-number gain) and indirectly caused by increased JAK2 signaling, has been associated with immunotherapy response against tumors with 9p amplification (56, 57). These results may apply to other tumors/sites and therapies, especially given the pivotal, broad role of JAK2 in cancer cell sensitivity to IFNγ, impaired antigen presentation, T-cell sensitivity, and evasion.

A wealth of functional data provided strong motivation for the generation of inhibitors that block the enzymatic activity of JAK as a new type of immunomodulatory medication. Preclinical disease models, including in arthritis, transplantation, graft-versus-host disease and other autoimmune conditions led to clinical trials. Moreover, the discovery of gain-of-function mutations of JAK2 in myeloproliferative neoplasms, including myelofibrosis, polycythemia vera, and essential thrombocytosis, provided additional impetus for developing JAK inhibitors (jakinibs) for the clinic (Table 3; refs. 58–67).

The first JAK inhibitor to receive approval was the JAK1/JAK2 inhibitor ruxolitinib (50, 61, 67). On the basis of the identification of gain-of-function JAK2 mutations in myelofibrosis, ruxolitinib was approved in 2011 for this indication. Subsequently, ruxolitinib has also been approved for steroid-refractory acute graft-versus-host disease, and a cream form for atopic dermatitis. More recently, fedratinib, a selective JAK2 inhibitor, was also approved for myelofibrosis (64).

Tofacitinib, a JAK1/JAK2/JAK3 inhibitor, is approved for RA, psoriatic arthritis, polyarticular course juvenile idiopathic arthritis, and ulcerative colitis (57, 58). Baricitinib is a JAK1/JAK2 inhibitor approved for RA and more recently for alopecia areata. It is also approved for the treatment of atopic dermatitis in Europe and has recently received emergency FDA approval, with and without the antiviral remdesivir, for the treatment of COVID-19 in hospitalized patients (62, 63). Upadacitinib has a degree of selectivity for JAK1 over JAK2 and is approved for the treatment of RA, psoriatic arthritis, and atopic dermatitis (63). Abrocitinb is also a selective JAK1 inhibitor approved for atopic dermatitis (64). Several other JAK inhibitors have been authorized for similar indications in Europe or Japan: filgotinib, peficitinib, delgocitinib, and oclacitinib (66).

Ongoing clinical trials are investigating the efficacy of jakinibs in a wide variety of additional conditions such as alopecia areata, ankylosing spondylitis, inflammatory bowel disease, dermatomyositis, interstitial lung disease, lupus, vasculitides, vitiligo, and myasthenia gravis. Finally, recent results suggest improvement in dermatologic abnormalities and in neurologic function in children with Aicardi–Goutieres syndrome (a genetic interferonopathy associated with severe disability and death; ref. 68).

JAK inhibitors carry an increased risk of infection as a side effect. The incidence of common infections such as upper respiratory tract, lower respiratory tract (including tuberculosis), and urinary tract infections are higher compared with the general population (69). In addition, for those JAK inhibitors used to treat chronic inflammatory conditions, the FDA requires warning about increased risk of cardiac events, cancer, and blood clots (70, 71).

Different components of the JAK/STAT pathway, such as receptors, JAK, STAT, and gene transcription factors crosstalk with other signaling pathways. These signaling cross talks play very important roles in pluripotency, differentiation, immune regulation, and tumorigenesis, which may be important for future therapeutic developments (72, 73). Specifically, components of the TGFβ-signaling pathway, which is involved in embryonic development and cell homeostasis, interact with components in the JAK/STAT pathway, and can either upregulate or downregulate the pathway (72, 73). SMAD proteins, the modulators of the TGFβ pathway, and STAT proteins often share the same transcription complex. One example is the STAT3 and SMAD1 complex linked by p300, which induces astrocyte differentiation.

The TGFβ-signaling pathway is modulated by the Notch pathway, which is involved in cell proliferation, differentiation, and cell death (72, 73). Components of the Notch pathway also crosstalk with the JAK/STAT pathway and have been studied in organ development in Drosophila. The Notch pathway suppresses JAK/STAT signals through interference of STAT translocation to the DNA domain (72). For example, in the development of the Drosophila central nervous system, Hes proteins (downstream effectors of Notch) facilitate JAK2 phosphorylation of STAT3 (72, 73).

The crosstalk between these pathways suggests that TGFβ and Notch pathway modulators may deserve consideration in the future when developing JAK pathway inhibitors.

JAK is a family of intracellular, non-receptor tyrosine kinase enzymes that convert cytokine-mediated signals via the JAK–STAT pathway into nuclear effects. Over 50 cytokines signal via the JAK/STAT pathway to coordinate hematopoiesis, immune function, and inflammation. The four JAK family members are: JAK1, JAK2, JAK3, and TYK2. The STAT family, which mainly act as transcription factors, is composed of seven members STAT1, STAT2, STAT3, STAT4, STAT5a, STAT5b, and STAT6. The cytokine signaling cascade requires only three elements (receptor, kinase, and transcription factor) to elicit a biologic impact. Each cytokine binds to a specific receptor, which promotes activation of specific JAK(s) and STAT(s); receptor-localized STATs are phosphorylated by JAK, which leads to their disassociation from the receptor and translocation to the nucleus, where they induce the expression of cytokine-responsive genes; signaling is switched off by a specific SOCS protein via a negative feedback loop.

JAK family members are associated with both germline mutations and polymorphisms that correlate with specific disease manifestations (Table 1; refs. 29–36). The autoimmune disease systemic lupus erythematosus has been linked to TYK2 polymorphisms. JAK3 germline mutations have been seen in severe combined immunodeficiency syndrome and in clear cell renal cell cancer. Pediatric acute lymphoblastic leukemia has been associated with germline TYK2 mutations, and hereditary thrombocytosis with JAK2 germline mutations.

Somatic gain-of-function mutations in JAK genes are predominantly associated with hematologic malignancies (Table 2; refs. 37–48). The somatic mutation JAK2 V617F, which constitutively activates JAK2 kinase, is a hallmark of myeloproliferative disorders, including polycythemia vera, essential thrombocytosis and primary myelofibrosis. JAK2 V617F is also found in the normal population and increases with aging as part of a phenomenon known as clonal hematopoiesis of indeterminate potential, which enhances risk for development of leukemia as well as for cardiac events and thromboembolism. JAK family somatic mutations are also observed in a subset of patients with a variety of T-cell malignancies—T-ALL, T-PLL, and Sezary syndrome—as well as in childhood B-cell precursory ALL, and acute megakaryoblastic leukemia (with and without Down syndrome) and in Castleman's disease. Finally, an activating TEL–JAK2 fusion protein has been reported in ALL and in atypical chronic myelogenous leukemia (42).

JAK2 copy-number loss, resulting in abrogation of JAK2 function, is also important. In particular, JAK2 gene loss is a pivotal driver of immune resistance and escape in cancer, and is associated with resistance to immunotherapy.

Because JAK family gain-of-function aberrant proteins are implicated in a spectrum of illnesses, JAK inhibitors have been deployed to treat these conditions. JAK inhibitors or jakinibs are now approved for a variety of medical problems including, but not limited to myeloproliferative disorders (ruxolitinib and fedratinib), graft-versus-host disease (ruxolitinib), RA (tofacitinib, baricitinib, and upadacitinib), psoriatic arthritis (tofactitinib and upadacitinib) atopic dermatitis (abrocitinib, upadacitinib and topical ruxolitinib), ulcerative colitis (tofacitinib), ankylosing spondylitis (tofacitinib, upadacitinib), and COVID-19 in hospitalized patients (baricitinib; Table 3; refs. 58–67), as well as most recently, baricitinib for alopecia (74). Future research will elucidate the potential of jakinibs in numerous other autoimmune illnesses, including, but not limited to alopecia areata, ankylosing spondylitis, inflammatory bowel disease, dermatomyositis, lupus, vasculitides, vitiligo and myasthenia gravis. Interestingly, recent observations suggest improvement in neurologic function and in skin abnormalities in children with Aicardi–Goutieres syndrome (a genetic interferonopathy resulting in severe disability and death). Interactions of JAK molecules with other signaling pathways such as TGFβ and notch may also need to be considered. Taken together, JAK inhibitors are transforming outcomes for many disorders, from cancer to autoimmunity and beyond.

R. Kurzrock reports grants, personal fees, non-financial support, and other support from Biological Dynamics, Boehringer Ingelheim, Debiopharm, Foundation Medicine, Genentech, Grifols, Guardant, Incyte, Konica Minolta, Medimmune, Merck Serono, Omniseq, Pfizer, Sequenom, Takeda, and TopAlliance; as well as consultant and/or speaker fees and/or advisory board for Actuate Therapeutics, AstraZeneca, Bicara Therapeutics, Biological Dynamics, Daiichi Sankyo, Inc., EISAI, EOM Pharmaceuticals, Iylon, Merck, NeoGenomics, Neomed, Pfizer, Prosperdtx, Roche, TD2/Volastra, Turning Point Therapeutics, X-Biotech; and reports an equity interest in CureMatch Inc., CureMetrix, and IDbyDNA; as well as serves on the Board of CureMatch and CureMetrix, and is a co-founder of CureMatch outside the submitted work. No disclosures were reported by the other authors.

This work was supported in part by NIH P30 CA023100 (to S.M. Lippman).